Published online before print July 14, 2003, doi: 10.1161/01.CIR.0000079222.91910.EE
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 2003;108:292.)
© 2003 American Heart Association, Inc.
Clinical Investigation and Reports |
Variability of Phase Shift Between Blood Pressure and Heart Rate Fluctuations
A Marker of Short-Term Circulation Control
Kára, MD
ek, MD, PhD
, MD
í Toman, MD, PhD
umbera, MD, PhDFrom the Institute of Scientific Instruments (J.H., P.J.), Academy of Sciences; St Anne’s University Hospital (T.K., M.S., M.N., R.P., J.T., J.S.); and the Faculty of Medicine (Z.N.), Masaryk University, Brno, Czech Republic; Mayo Clinic (T.K., W.K.S., V.K.S.), Rochester, Minn; Chelsea & Westminster Hospital (D.P.F.), London; Colchester General Hospital (L.C.D.), Colchester; and the National Heart & Lung Institute (A.J.S.C.), London, UK.
Reprint requests to Josef Halámek, PhD, Institute of Scientific Instruments, AS CR, Královopolská 147, 612 64 Brno, Czech Republic. E-mail josef{at}isibrno.cz
Received November 12, 2002; revision received April 17, 2003; accepted April 18, 2003.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background— We postulated that the variability of the phase shift between blood pressure and heart rate fluctuation near the frequency of 0.10 Hz might be useful in assessing autonomic circulatory control.
Methods and Results— We tested this hypothesis in 4 groups of subjects: 28 young, healthy individuals; 13 elderly healthy individuals; 25 patients with coronary heart disease; and 19 patients with a planned or implanted cardioverter-defibrillator (ICD recipients). Data from 5 minutes of free breathing and at 2 different, controlled breathing frequencies (0.10 and 0.33 Hz) were used. Clear differences (P<0.001) in variability of phase were evident between the ICD recipients and all other groups. Furthermore, at a breathing frequency of 0.10 Hz, differences in baroreflex sensitivity (P<0.01) also became evident, even though these differences were not apparent at the 0.33-Hz breathing frequency.
Conclusions— The frequency of 0.10 Hz represents a useful and potentially important one for controlled breathing, at which differences in blood pressure–RR interactions become evident. These interactions, whether computed as a variability of phase to define stability of the blood pressure–heart rate interaction or defined as the baroreflex sensitivity to define the gain in heart rate response to blood pressure changes, are significantly different in patients at risk for sudden arrhythmic death. In young versus older healthy individuals, only baroreflex gain is different, with the variability of phase being similar in both groups. These measurements of short-term circulatory control might help in risk stratification for sudden cardiac death.
Key Words: baroreceptors • respiration • death, sudden
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Circulation control is governed primarily by feedback systems.1,2 Circulation control by the baroreflex, for example, could be characterized as an input signal represented by systolic blood pressure and an output signal represented by RR interval. The relation between input and output signals in a feedback system is defined by the transfer function. Attributes of the transfer function include gain and phase. Gain characterizes the relation between the amplitudes of the input and output signals. In the case of the baroreflex, gain might correspond, for example, to the magnitude of change in RR interval for a given change in blood pressure. Phase characterizes the delay between input and output signals. In the case of the baroreflex, the phase relation is described by quantifying the phase shift between systolic blood pressure and RR interval.
In feedback systems, phase determines the stability.3 Stability is a basic prerequisite of any control system and is more important than gain. A high value of baroreflex gain, in the absence of a more detailed phase-shift analysis, might provide, on its own, little information on the stability of short-term circulation control. If the phase of transfer function is not steady, from the point of view of hemodynamic stability, then it is conceivable that any benefits of increased gain of the baroreflex control of heart rate might be less apparent. The stability of the phase shift between systolic blood pressure (SBP) and RR interval (RR) fluctuation in the low-frequency (LF) band is best described graphically by phase of the instantaneous transfer function.3
Breathing, blood pressure, and RR have powerful and physiologically important interactions.4–8 At higher breathing frequency (eg, 0.33 Hz), the frequency of breathing-related perturbations of the blood pressure–RR interaction lies outside the LF band. Therefore, at this high frequency, any breathing-related changes in measurements of the gain and stability of the blood pressure–RR interaction are diminished. At the lower frequency of 0.10 Hz, the maximum gain of the baroreflex is expected to be manifest.9,10 At this "natural harmonic" of the arterial baroreflex, breathing-induced changes in blood pressure and RR interval are most readily translated into blood pressure–mediated changes in RR intervals. Thus, differences in gain and stability between individuals might be more readily evident.
The overall goal of these studies was to examine the dynamic behavior of short-term circulatory control in healthy old and young subjects and in patients with cardiovascular disease. The coherence between SBP and RR oscillations in the LF band was analyzed by controlled breathing that was used to perturb the baroreflex feedback system. We tested the hypothesis that a 0.10-Hz breathing frequency would most clearly identify differences in baroreflex gain and stability between young and older subjects, as well as identify those patients with cardiovascular disease at greatest risk for sudden cardiac death.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
We measured short-term circulatory control (baroreflex gain and variability of phase) in 4 groups (Table 1): (1) young, healthy volunteers (n=28; 6 female); (2) older, healthy volunteers (n=13; 2 female); (3) patients with angiographically documented coronary heart disease (CHD) (n=25; 3 female), without any life-threatening arrhythmias. Ejection fraction (mean±SD) was 47±9% (range, 30% to 65%). Functional class assessment showed 5, 8, 11, and 1 subjects in New York Heart Association (NYHA) classes I, II, III, and IV, respectively. Twenty patients had a previous myocardial infarction, 18 had hypertension, and 6 had diabetes. Medications were as follows: aspirin (all); ß-blockers (n=24); angiotensin-converting enzyme inhibitors (n=23); calcium antagonists (n=13); diuretics (n=5); nitrates (n=19); and lipid therapy (n=19). After more than 1 year of follow-up for the study, no patients had any life-threatening arrhythmias or sudden death episodes. Group (4), implanted cardioverter-defibrillator (ICD) recipients, were patients with an ICD (n=14; 1 female) and candidates scheduled for ICD placement (n=5; 2 female) for life-threatening ventricular tachycardia or ventricular fibrillation. The time of ICD implantation before the study was up to 3 years. Fourteen patients had coronary artery disease, 2 had dilated cardiomyopathy, and 3 had no structural heart disease. The defibrillator had been previously activated in 6 patients to terminate ventricular tachycardia or fibrillation. Ejection fraction was 37±13% (range, 18% to 65%). Functional class assessment showed 2, 7, 9, and 1 patients in NYHA classes I, II, III, and IV, respectively. Medications were aspirin (n=1); ß-blockers (n=5); angiotensin-converting enzyme inhibitors (n=10); calcium antagonists (n=1); diuretics (n=10); nitrates (n=7); lipid therapy (n=5); and antiarrhythmics (n=16).
|
Continuous, 5-minute measurement of SBP and RR, with <4 extrasystoles, was performed under strictly standardized conditions during free (spontaneous) breathing and controlled breathing at 6 breaths per minute (0.10 Hz) and 20 breaths per minute (0.33 Hz), in the supine position. Detailed training of the patients preceded each measurement. The interval between measurements was 5 minutes.
We recorded ECG, blood pressure (Finapres-2300, Ohmeda), and breathing (Spirometer ANNAlab SM-1, St Anne’s University Hospital, Brno, Czech Republic). The frequency and depth of breathing were optically regulated by means of a light-emitting-diode indicator. Signals were digitized through an A/D 16-bit converter with a sampling frequency of 500 Hz. All measurements were completed in the Laboratory for Research of Circulation Control of St Anne’s University Hospital. The study was approved by the Institutional Review Committee of St Anne’s University Hospital. Each subject gave informed consent.
Statistical Analysis
All measurements and results are presented as mean±SD. Differences between groups were assessed by ANOVA. The following key measurements were obtained and compared between groups: (1) baroreflex sensitivity of heart rate computed both by cross-spectral algorithm and as the 
-index.11–13 (2) mean phase of instantaneous transfer function (
mean) and fluctuation of phase (STD; SD of instantaneous phase) to describe the phase conditions in the feedback loop and the coherence of regulated signals; (3) heart rate variability as (a) LF power (LFRR), 0.05 to 0.15 Hz; (b) high-frequency power (HFRR), 0.15 to 0.4 Hz; (c) normalized RR variability (NLFRR-LF power divided by power in LF and HF band); (d) LF/HFRR(14–16); and (e) total variability (RRtot).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In comparisons between young and elderly healthy subjects, during free breathing and controlled breathing at 20 breaths per minute (0.33 Hz), neither cross-spectral nor
-index computations of baroreflex gain, nor phase of instantaneous transfer function and fluctuation of phase was significantly different (Figure 1, Table 2). However, at 6 breaths per minute (0.10 Hz), differences in baroreflex gain measured both cross-spectrally and as the -index were very clearly evident. However, there was no evidence for differences in variability of phase between young and elderly healthy people, suggesting that age did not affect the stability of the blood pressure–RR interaction, even though the gain of this interaction was significantly different in old versus young subjects (Figure 1). Comparison of the effects of breathing frequency on fluctuation of phase revealed that for all groups, the fluctuation of phase was lower during the 0.10-Hz breathing compared with 0.33-Hz breathing (Figure 1).
|
|
Both the CHD and ICD populations were comparable in terms of age and medications. The high prevalence of postmyocardial infarction patients in the CHD group was consistent with a high-risk category, even though these subjects had had no events for at least 1 year of follow-up since completion of this study. Despite the similarities between the CHD and ICD groups, with the obvious exception of a defined, high risk for sudden death in the ICD recipient group, no differences in cross-spectral and -index computations of baroreflex gain were evident in the 2 groups, either during free breathing or at a breathing frequency of 20 breaths per minute (0.33 Hz). Mean phase of instantaneous transfer function was also not different between these patient populations. However, at a breathing frequency of 6 breaths per minute (0.10 Hz), cross-spectral and -index computations of baroreflex gain were clearly manifest. Mean phase of instantaneous transfer function was still comparable between groups. However, at all breathing frequencies, the fluctuation of phase, the key index of variability of phase, was significantly different between the CHD and ICD groups (Tables 2 and 3
). Figure 2 shows graphical representations of the instantaneous transfer function at a 0.10-Hz breathing frequency. Figure 3 shows the correlation between baroreflex sensitivity and fluctuation of phase in individual subjects, documenting the clear difference in the ICD recipients with high risk for ventricular fibrillation.
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
These studies demonstrate that ICD recipients have unstable phase control, which is not attributable to the normal aging process or CHD. Phase control, unlike baroreflex sensitivity, is not altered with normal aging. In healthy subjects, differences in baroreflex gain, whether computed as cross-spectral measurements or as the
-index, are clearly evident in young versus old subjects, such that the baroreflex gain is greater in younger subjects.17 It is important that at a breathing frequency of 0.33 Hz, significant differences in baroreflex gain might not be evident. Measurements of variability of phase, an index of the stability of the blood pressure–RR interaction, are similar in young and old people at both breathing frequencies. Thus, variability of phase, or the stability of the blood pressure–RR interaction, does not appear to be influenced by age. Variability of phase, however, appears to be affected significantly by the presence of an increased risk for sudden cardiac death. In the ICD recipients, significant differences in comparison with the other 3 groups are evident for fluctuation of instantaneous phase between SBP and heart rate variability, and these differences are also clearly evident at all breathing frequencies. Nevertheless, it is interesting that this measurement of the instability of the blood pressure–RR interaction is strikingly lower for all groups during the 0.10-Hz breathing frequency compared with spontaneous breathing or the 0.33-Hz breathing frequency (P<0.0001). These findings might have direct implications for understanding interventions that relate to improving cardiovascular autonomic control and might have relevance to recent work by Bernardi and colleagues,18–20 who showed that changes in breathing frequency might have important influences on neural circulatory control in healthy subjects and patients with heart failure.
The normalized LF power of RR variability in the ICD recipients might also provide important insights.21–23 The RR variability in the ICD recipients is low and spread across all frequency bands. At a controlled breathing frequency of 0.10 Hz, the LF power of RR variability increases only slightly in the ICD patients, and the RR variability remains distributed over all frequency bands. This is in striking contrast to the other groups, in whom the coupling between breathing and RR variability is much higher and in whom RR variability increases very clearly at a breathing frequency of 0.10 Hz when the dominant expected power of RR variability would be in the LF band. This absence of entrainment of RR variability by breathing might relate to both the gain and stability of baroreflex function in the ICD recipients. A diminished baroreflex gain would reduce the ability of breathing-induced changes in pressure to entrain the appropriate baroreflex-mediated changes in RR interval. Loss of stability of the blood pressure–RR interactions would further contribute to an inability to maintain power of RR oscillations within the LF band in patients with unstable blood pressure- RR control systems.
Limitations of this study include, first, that the frequency range of short-term circulation-control assessment that we used was the classic range, where we anticipated a baroreflex frequency bandwidth. It is possible that a more clearly defined range for operation of the baroreflex would provide even clearer data. Second, strictly standardized conditions before and during these measurements are essential for obtaining reliable results. Distractions or disturbances during measurements might generate phase peaks. Third, drug therapy in our patient population might have influenced between-group comparisons. In mitigation, there were clear differences between the CHD group and the ICD group, even though both groups were taking cardiovascular medications. Furthermore, the between-subject effects of the 0.1-Hz and 0.33-Hz breathing frequency were similar across patients and normal subjects.
In summary, we have examined, in young and old healthy subjects and in patients with cardiovascular disease with and without a high risk for sudden cardiac death, measurements of baroreflex gain and circulatory stability at a spontaneous breathing frequency and at breathing frequencies of 0.33 Hz and 0.10 Hz. At a 0.10-Hz breathing frequency, the presumed natural resonating frequency of the baroreflex, differences in baroreflex gain in young and old people are readily apparent, even though these differences are not significant during spontaneous breathing and at a breathing frequency of 0.33 Hz. Circulatory stability, measured by variability of phase, is not different in old versus young people. However, in patients with cardiovascular disease at high risk for sudden cardiac death, measurements of circulatory stability, as computed by fluctuation of phase, are strikingly different in the sudden cardiac death risk group compared with all other groups. Again, reduced baroreflex gain in these patients at high risk for sudden death is apparent only at the 0.10-Hz breathing frequency. Even at a 0.10-Hz breathing frequency, the magnitude of LF power of the RR oscillations is attenuated in the high-risk ICD group. This might reflect both the diminished baroreflex gain as well as the circulatory instability24,25 in the ICD recipients.
Excitation of short-term circulatory control by controlled breathing at a frequency of 0.10 Hz might represent a simple and convenient, noninvasive approach to amplify the characteristics of cardiovascular control and to eliminate excess noise. Without adequate excitation, fluctuations of SBP and RR, corresponding to circulatory control mechanisms, might be observed by random variation and extraneous noise. This might explain the absence of clear differences in baroreflex gain at breathing frequencies of 0.33 Hz. Important information about both neural circulatory control and risk for arrhythmia might be provided by combined analysis of reflex gain26–30 and variability of the phase. Differences in baroreflex gain can occur in the absence of any changes in variability of phase, as is evident in young versus elderly healthy subjects. Conversely, changes in variability of phase consistent with circulatory instability might be manifest even in the absence of measurable differences in baroreflex gain, as is evident in the ICD recipients studied during spontaneous breathing and at a breathing frequency of 0.33 Hz. Both baroreflex gain and variability of phase might provide important and independent information about circulatory control and stability and might provide important additional information regarding stratification of risk of sudden cardiac death.
|
Acknowledgments |
|---|
The work was supported by grants 102/00/1262 and 102/02/1339 (Grant Agency of the Czech Republic), CEZ: J07/98:141100004 (Ministry of Education, Youth, and Sports, Czech Republic) and internal grant 2001/06091938 (St Anne’s University Hospital). Drs Kára, Shen, and Somers are supported by National Institutes of Health grants HL61560, HL 65176, HL 70302, and MO1-RR00585. The authors acknowledge Jan N
mec, MD, and Zdenk Placheta, MD, DrSc, for their valuable comments. |
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Furlan R, Jacob G, Palazzolo L, et al. Sequential modulation of cardiac autonomic control induced by cardiopulmonary and arterial baroreflex mechanisms. Circulation. 2001; 104: 2932–2937.
2. Legramante JM, Raimondi G, Massaro M, et al. Investigating feed-forward neural regulation of circulation from analysis of spontaneous arterial pressure and heart rate fluctuations. Circulation. 1999; 99: 1760–1766.
3. Jurák P, Halámek J, Kára T, et al. Immediate baroreflex sensitivity: new approach to this parameter. Med Biol Eng Comput. 1999; 37: 496–497.
4. Pitzalis MV, Mastropasqua F, Massari F, et al. Effect of respiratory rate on the relationships between RR interval and systolic blood pressure fluctuations: a frequency-dependent phenomenon. Cardiovasc Res. 1998; 38: 332–339.
5. Piepoli M, Sleight P, Leuzzi S, et al. Origin of respiratory sinus arrhythmia in conscious humans: an important role for arterial carotid baroreceptors. Circulation. 1997; 95: 1813–1821.
6. Diehl RR, Linden D, Berlit P. Determinants of heart rate variability during deep breathing: basic findings and clinical applications. Clin Auton Res. 1997; 7: 131–135.[CrossRef][Medline] [Order article via Infotrieve]
7. Frederiks J, Swenne CA, TenVoorde BJ, et al. The importance of high-frequency paced breathing in spectral baroreflex sensitivity assessment. J Hypertens. 2000; 18: 1635–1644.[CrossRef][Medline] [Order article via Infotrieve]
8. Pentilla J, Helminen A, Jartti T, et al. Time domain, geometrical and frequency domain analysis of cardiac vagal outflow: effects of various respiratory patterns. Clin Physiol. 2001; 21: 365–376.[CrossRef][Medline] [Order article via Infotrieve]
9. Linden D, Diehl RR. Estimation of baroreflex sensitivity using transfer function analysis: normal values and theoretical considerations. Clin Auton Res. 1996; 6: 157–161.[CrossRef][Medline] [Order article via Infotrieve]
10. Davies LC, Francis DP, Jurák P, et al. Reproducibility of methods for assessing baroreflex sensitivity in normal controls and in patients with chronic heart failure. Clin Sci. 1999; 97: 515–522.[Medline] [Order article via Infotrieve]
11. Baselli G, Cerutti S, Civardi S, et al. Spectral and cross-spectral analysis of heart rate and arterial blood pressure variability signals. Comput Biomed Res. 1986; 19: 520–534.[CrossRef][Medline] [Order article via Infotrieve]
12. Parati G, Saul JP, Di Rienzo M, et al. Spectral analysis of blood pressure and heart rate variability in evaluating cardiovascular regulation: a critical appraisal. Hypertension. 1995; 25: 1276–1286.
13. Di Rienzo M, Parati G, Tordi R, et al. Signal processing techniques for the evaluation of spontaneous baroreflex. Med Biol Eng Comput. 1999; 37: 402–403.
14. Pagani M, Lombardi F, Guzzetti S, 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: 178–193.
15. Pagani M, Lombardi F, Guzzetti S, et al. Power spectral density of heart rate variability as an index of sympatho-vagal interaction in normal and hypertensive subjects. J Hypertens. 1984; 2: S383–S385.
16. Iellamo F, Pizzinelli P, Massaro M, et al. Muscle metaboreflex contribution to sinus node regulation during static exercise: insights from spectral analysis of heart rate variability. Circulation. 1999; 100: 27–32.
17. Matsukawa T, Sugiyama Y, Mano T. Age-related changes in baroreflex control of heart rate and sympathetic nerve activity in healthy humans. J Auton Nerv Syst. 1996; 60: 209–212.[CrossRef][Medline] [Order article via Infotrieve]
18. Bernardi L, Spadacini G, Bellwon J, et al. Effect of breathing rate on oxygen saturation and exercise performance in chronic heart failure. Lancet. 1998; 351: 1308–1311.[CrossRef][Medline] [Order article via Infotrieve]
19. Spicuzza L, Gabutti A, Porta C, et al. Yoga and chemoreflex response to hypoxia and hypercapnia. Lancet. 2000; 356: 1495–1496.[CrossRef][Medline] [Order article via Infotrieve]
20. Bernardi L, Porta C, Spicuzza L, et al. Slow breathing increases arterial baroreflex sensitivity in patients with chronic heart failure. Circulation. 2002; 105: 143–145.
21. Malliani A, Lombardi F, Pagani M, et al. Power spectral analysis of cardiovascular variability in patients at risk for sudden cardiac death. J Cardiovasc Electrophysiol. 1994; 5: 274–286.[Medline] [Order article via Infotrieve]
22. Bigger JT Jr, Fleiss JL, Steinman RC, et al. Frequency domain measures of heart period variability and mortality after myocardial infarction. Circulation. 1992; 85: 164–171.
23. Bigger JT Jr, Fleiss JL, Rolnitzky LM, et al. Frequency domain measures of heart period variability to assess risk late after myocardial infarction. J Am Coll Cardiol. 1993; 21: 729–736.[Abstract]
24. Huikuri HV, Seppanen T, Koistinen MJ, et al. Abnormalities in beat-to-beat dynamics of heart rate before the spontaneous onset of life-threatening ventricular tachyarrhythmias in patients with prior myocardial infarction. Circulation. 1996; 93: 1836–1844.
25. Nemec J, Hammill SC, Shen WK. Increase in heart rate precedes episodes of ventricular tachycardia and ventricular fibrillation in patients with implantable cardioverter defibrillators: analysis of spontaneous ventricular tachycardia database. Pacing Clin Electrophysiol. 1999; 22: 1729–1738.[CrossRef][Medline] [Order article via Infotrieve]
26. La Rovere MT, Bigger JT Jr, Marcus FI, et al. Baroreflex sensitivity and heart-rate variability in prediction of total cardiac mortality after myocardial infarction: ATRAMI (Autonomic Tone and Reflexes After Myocardial Infarction) Investigators. Lancet. 1998; 351: 478–484.[CrossRef][Medline] [Order article via Infotrieve]
27. Schwartz PJ, La Rovere MT. ATRAMI: a mark in the quest for the prognostic value of autonomic markers: Autonomic Tone and Reflexes After Myocardial Infarction. Eur Heart J. 1998; 19: 1593–1595.
28. Malik M, Camm AJ, Janse MJ, et al. Depressed heart rate variability identifies postinfarction patients who might benefit from prophylactic treatment with amiodarone: a substudy of EMIAT (The European Myocardial Infarct Amiodarone Trial). J Am Coll Cardiol. 2000; 35: 1263–1275.
29. La Rovere MT, Pinna GD, Hohnloser SH, et al. Baroreflex sensitivity and heart rate variability in the identification of patients at risk for life-threatening arrhythmias: implications for clinical trials. Circulation. 2001; 103: 2072–2077.
30. Schwartz PJ, La Rovere MT, Vanoli E. Autonomic nervous system and sudden cardiac death: experimental basis and clinical observations for post-myocardial infarction risk stratification. Circulation. 1992; 85 (suppl I): I-77–I-91.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  K. Narkiewicz, P. van de Borne, N. Montano, D. Hering, T. Kara, and V. K. Somers Sympathetic Neural Outflow and Chemoreflex Sensitivity Are Related to Spontaneous Breathing Rate in Normal Men Hypertension, January 1, 2006; 47(1): 51 - 55. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Baba, T. Yoshikawa, Y. Fukuda, T. Sugiyama, M. Shimada, M. Akaishi, K. Tsuchimoto, S. Ogawa, and M. Fu Autoantibodies against M2-muscarinic acetylcholine receptors: new upstream targets in atrial fibrillation in patients with dilated cardiomyopathy Eur. Heart J., July 1, 2004; 25(13): 1108 - 1115. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||