CC BY-NC-ND
0Review Article
Circadian Distribution and Autonomic Tone Modulation in Paroxysmal Atrial Fibrillation
Teruhisa Tanabe
Department of Cardiology, School of Medicine, Tokai University
Understanding the role played by autonomic tone changes in the initiation of paroxysmal atrial fibrillation (PAF) may be helpful in preventing the occurrence of the arrhythmia. The purpose of this review article is to discuss the relationship between sympathovagal imbalance and the initiation of PAF from past literature and our data regarding circadian distribution and heart rate variability (HRV) analysis in PAF. PAF in non-structural heart disease (NHD) frequently occurs during evening and night hours with a longer duration of each episode, and its initiation is directly associated with an increase in vagal tone. According to our data of HRV analysis in NHD, HF power (indicator of vagal tone) showed a reduction 15 to 30 minutes before PAF, followed by a sharp increase immediately before the onset, suggesting a primary increase in sympathetic tone followed by a marked changes toward vagal predominance. Very few studies have been reported on PAF onset or the modulation of autonomic tone in structural heart disease (SHD). Some studies showed no significant differences in the circadian variation of the onset between NHD and SHD. However, we demonstrated a triphasic circadian pattern with maximum peaks in early morning, late afternoon, and night in SHD, suggesting an association with not only a vagally induced origin but also a sympathecally induced or stress triggered origin in PAF onset in SHD. The occurrence of PAF greatly depends on modulation of autonomic tone and is extremely complicated. Further studies are required to clarify the relationship between the occurrence of PAF and autonomic modulation. (J Arrhythmia 2008; 24: 122-132)
„ . PAF, Circadian variation, Autonomic changes, Structural heart disease, Non-structural heart
Key words: °
disease
Introduction
Circadian distribution has been reported in some cardiovascular disorders such as acute myocardial infarction,1-3 sudden cardiac death,4,5) transient myocardial ischemia6,7 and stroke,8 with a predominance of incidents occurring in the early morning.
One reason for this is considered to be related to an abrupt change from a predominantly vagal tone to a dependent sympathetic tone in the early morning. Ventricular premature contractions and supraven-tricular premature contractions have been known to demonstrate circadian fluctuation patterns with high reproducibility in each patient; daytime, nighttime and both day- and nighttime onset, and irregular
Address for correspondence: Teruhisa Tanabe MD, Department of Cardiovascular Medicine, School of Medicine, Tokai University, Isehara 259-1193, Japan. Tel: +81-463-93-1121 Fax: +81-463-93-6679 E-mail: ttanabe@is.icc.u-tokai.ac.jp
occurrence.9-11' There are also many reports regarding the circadian distribution of the occurrence of episodes of paroxysmal supraventricular tachycardia (PSVT).12-19)
In recent years, heart rate variability (HRV) analysis has enabled us to provide a quantitative evaluation of sympathovagal interaction in patients with cardiac disorders. Reduced vagal and augmented sympathetic activities have been increasingly linked to life-threatening ventricular arrhythmias, sudden cardiac death, total mortality after myocardial infarction, and congestive heart fail-ure.20-23) Alterations in sympathovagal balance with reduced vagal and increased sympathetic outflow to the heart are also reported in patients with advanced hypertrophic and dilated cardiomyop-athy.24)
Atrial fibrillation is a common arrhythmia with or without structural heart disease.25) Especially, initiation of paroxysmal atrial fibrillation (PAF) in patients with non-structural heart disease (NHD) is reportedly associated with specific circadian distri-bution13,15-18,26-30) and with changes of sympathetic or vagal tone seen clinically31-33 or with HRV analysis.28,34-38)
In this article we reviewed the circadian distribu-
tion of PAF and the effect of autonomic tone modification on PAF initiation.
Holter monitoring and atrial fibrillation
Holter monitoring is an indispensable tool for detection, diagnosis, or management of arrhythmia. In order to learn which clinical and basic uses of Holter monitoring have been favored by cardiologists around the world, a search for 'Holter monitoring' was done from literature included in PUBMED/ MEDLINE.39) The 10 most prevalent topics from a search for 'Holter monitoring' for literature published between January 2002 and March 2007 are shown in Figure 1. The most frequently observed subject was 'heart rate variability (HRV),' which was found in 182 out of 1,885 articles followed by 'atrial fibrillation' in 99 articles. Thus, Holter monitoring is frequently used to investigate HRV and atrial fibrillation.
Figure 2 shows the 10 most prevalent topics obtained from a search for 'Holter monitoring and atrial fibrillation' between January 2000 and March 2007 in articles included in PUBMED/ MEDLINE.39) Among them, the seventh most prevalent topic was 'autonomic nervous system'.
(99) Atrial fibrillation
(77) Cardiac sudden death, syncope
(71) During or after surgery
(69) Event recorder, trans-telephonic ECG
(65) QT interval, QT variability, QT/RR relation
(62) CAD except for MI
(60) Catheter ablation, cardioversion, other techniques
(60) Children including newborns, infant and pupils
(50) Innovation or development of Holter system
nos. of articles
Figure 1 Top 10 topics among 1885 articles from a PUB/MED search for 'Holter monitoring' between Jan 1, 2002 and March 31, 2007.
HRV: heart rate variability, CAD: coronary artery disease, MI: myocardial infarction
25 50 75 100
1 (74) Long-term management of AF nos. of articles
2 (34) Evaluation of antiarrhythmic drugs
3 (24) Anticoaguration
4 (15) Rhythm and rate control
5 (14) Atrial pacing
6 (6) Signal averaged ECG on 'p' waves
7 (4) Autonomic nervous system
8 (3) Highbrid treatment of pacing and drugs
9 (2) Evaluation after thromboembolic stroke
10 (2) Atrial size
Figure 2 The 10 most prevalent topics among 618 articles in PUB/MED search for 'Holter monitoring and atrial fibrillation' between Jun 1, 2000 and March 31, 2007. AF: atrial fibrillation.
Circadian distribution of supraventricular tachyarrhythmias
1. Paroxysmal supraventricular tachycardia PSVT, PAF and atrial flutter are well known supraventricular tachyarrhythmias with episodes of paroxysms. Table 1 shows review literature investigating circadian variation in the onset of supra-ventricular tachyarrhythmias including PAF and PSVT. Irwin et al.14) observed a peak incidence of PSVT in the afternoon and evening with a maximum at 4:00 pm. Regarding the PSVT, there is relatively broad consensus that episodes occur mainly during the afternoon.17,18) However, there are studies demonstrating a double-peak with a peak occurring with PSVT in the morning in addition to the afternoon peak.16,19) Plasma concentration was higher during the daytime (8:00 am to 6:00pm) than at night in patients with essential hypertension.40) The reason for this high incidence of PSVT mainly in the afternoon or early morning may be based on an acceraration in electrophysiologic conduction of the atrioventricular node induced by adrenergic preva-lence.41) As a matter of fact, the abolition of the
morning peak induced by ^-blockers is consistent with this hypothesis.16)
2. Paroxysmal atrial fibrillation
There are many studies on circadian distribution of the onset of PAF.15-18,26-30) Some of them have demonstrated a double-peak occurrence of PAF with morning and evening,16,29 whereas the otherers showed peaks during midnight and in the morning,17) or after lunch and at midnight.27) Also, there are studies demonstrating a much higher occurrence during the day than at night,26,28) triple peaks—noon to 2:00pm, 6:00 pm to 2:00 am and 4:00 am to 6:00 am,30) and no circadian distribution.18) The data for PAF showing a peak incidence during the night are completely different from those for PSVT in which episodes mainly occurred in the afternoon. The mechanisms responsible for this difference are not clear.
Experimentally, atrial fibrillation can be initiated by application of acetylcholine to the localized region of the auricular surface.42) Clinically, in patients with frequent PAF episodes, especially with NHD, the onset of paroxysms is often associated with an increase in vagal tone.31,32) Therefore, a
Table 1 Literatures reviewed from investigation of the circadian distribution of proxysmal atrial fibrillation atrial (PAF) and paroxysmal supraventricular tachycardia (PSVT).
Author Publish (year) Nos. of patients (men/women) Mean age (y.o) Arrhythmia (Nos. of episodes) Underlying HD (Nos. of patients) Drugs symptomatic (S) or asymptomatic (A) Peak time of incidence
Irwin JM, et al. 1988 52 (?) 46.2 ± 15.7 PSVT (52) presence non S, telephone contact or regular clinic visits PSVT: afternoon and evening, with peak at 4 pm
Kupari M, et al. 1990 251 (170/81) 49 PSVT (69), PAF (152) presence (127) absence (82) AAD, ^-blocker S, emergency room PAF: double-peakingwith evening and morning. PSVT: morning
Rawles JM, et al. 1990 72 (?) 72 (34-93) PAF (139) not described AAD, ^-blocker, Digoxin S & A, Holter monitoring PAF: more often by day than by night
PAF: with a peak between
Rostagno C, et al. 1993 1,118 (?) not described PSVT (348), PAF (726) presence, absence not described S, mobile coronary care unit midnight and 2am. With secondary peak in the morning (8 to 9 am). PSVT: during the daytime
Clair WK, et al. 1993 150 (79/71) not described PSVT (92), PAF (34) presence, absence drug free S, event ECG recorder and telephone transmission PAF: no circadian distribution. PSVT, with peak at approximately at 6 pm.
Yamashita T, et al. 1997 150 (120/30) 61.6 ± 12.8 407 PAF (407) absence drug free S & A, Holter monitoring PAF: with a double peak after lunch and at midnight
PAF: the most common onset
S & A, Holter monitoring time between 4:00 am and
Huang JL, et al. 1988 57 (34/23) 66 ± 22 PAF (121) presence (27) absence (30) not described 4:00pm (40/63). No different circadian distribution pattern between vagal and sympathetic type PAF.
PSVT: with a peak during day-
Lee SH, et al. 1999 105 (?) not described PAF (498) not described drug free S & A, Holter monitoring time, especially with peaks at 8:00am to 9:00am, 12:00pm to 1:00 pm, and 5:00 to 6:00 pm.
AAD S, event PAF: with a double peak with
Viskin S, et al. 1999 3,343 (1,872/1,471) 74.3 ± 12 PAF (9,989) presence, absence and other drugs ECG recorder and telephonic transmission a significant increase in the morning and and a second rise in evening
Vincenti A, et al. 2005 90 (58/32) 67.7 (34-93) PAF (233) presence (28) not described S & A, Holter PAF: triple peaks with noon to 2:00pm, 6:00pm to 2:00am,
monitoring and 4:00am to 6:00am.
subset of PAF has the increase in vagal tone as a partial mechanism since the above periods with a peak incidence of PAF during the night may be characterized by predominance of vagal tone.
However, most of the past studies vary greatly as to the circadian distribution of the onset of PAF, as described previously. Their weakness is that they do not measure the true circadian variation of PAF because they included only patients with symptomatic episodes,16-18,29 those with unusual circumstances such as emergency room16) or mobile coronary care unit,17) those taking antiarrhythmic drugs including y8-blockers,16,26,29) those with event ECG recorder or
transtelephonic transmission of ECG,18,29) and those with both structural heart disease (SHD) and NHD.16-18,28) Furthermore, some studies did not describe whether the patients were on antiarrhythmic drugs including y6-blockers,19,21,23,29) although this is a very important aspect.
Nevertheless, only one paper, which was published by Yamashita et al.,27) precisely described the patient characteristics of NHD, absence of drug therapy and inclusion of both symptomatic and asymptomatic subjects. They analyzed the circadian variation not only at the onset of PAF but also the hourly total duration, maintenance and termination
of the arrhythmia from 150 Holter recordings with a total of 407 episodes of PAF in a drug-free state with NHD. They demonstrated non-uniformity of hourly total duration across the 24 hours, reflecting a non-random occurrence of PAF throughout the day, with an increase at night with a peak at about midnight and a decrease in the morning with a nadir at about 11:00 am. The incidence of onset was also non-uniformly distributed throughout the 24-hour period because it showed a double peak with increases after lunch and at midnight. However, the probability of onset could not be fitted to a single- or double-harmonic curve, although the incidence had a weak double-peaked diurnal rhythm (Figure 3A). In contrast, they observed significant circadian variations for both the maintenance period and termination of PAF: a sharp decline in the morning to a nadir at about 10:00 am and then a gradual increase after lunch in the maintenance period and a double-curve with peaks at 11:00 am to 12:00 pm and 10:00 pm to 11:00 pm. The authors suggested that the maintenance and termination of PAF showed more distinct circadian variations compared with the onset (Figure 3B, C).
There are very few reports which compare the circadian distribution of the onset of PAF between SHD and NHD. Kupari et al.16) and Rostagno et al.17) described that no significant differences in the circadian variation of the onset of PAF existed between the 2 groups. However, they included only patients with symptomatic episodes who were admitted to hospital in unusual circumstances such as emergency room or mobile coronary care unit, and took antiarrhythmic drugs including fi-blockers.
We conducted a study of the circadian distribution of the hourly total duration, onset, maintenance and termination of PAF using Holter monitoring method in a drug-free state, which was similar to the protocol of Yamashita et al.27) comparing SHD and NHD patients. As a result, the circadian distribution at onset of PAF throughout 24 hours showed a marked difference between the 2 groups. The SHD group showed a triphasic circadian pattern with maximum peaks in the late afternoon, between late evening and night, and in the early morning, whereas the NHD group had a single harmonic peak between late evening and midnight (Figure 4). The coefficients of the two groups were significant (p < 0.05 in SHD and p < 0.01 in NHD). The circadian distribution of the probability of termination of PAF in SHD was demonstrated with a peak around noon and with a nadir at around midnight with a significant difference in the distribution (p < 0.05) (Figure 5). On the
A. Probability of Onset
0 2 4 6 8 10 12 14 16 18 20 22
B. Probability of Maintenance
ftnfffl
0 2 4 6 8
I I I I T HT T*T
10 12 14 16 18 20 22
0.6 0.5 0.4 0.30.2 0.1 0
C. Probability of Termination
0 2 4 6 8 10 12 14 16 18 20 22 Clock Time (hours)
Figure 3 Hourly probability of paroxysmal atrial fibrillation episodes (A: onset, B: maintenance, C: termination) (from Yamashita T et al: Circadian variation of paroxysmal atrial fibrillation. Circulation 1997; 96: 1537-41 Reference No. 27).
other hand, in the distribution in the NHD group, a weak double-peaks at around 6:00am-11:00am and
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1 2 3 4 5
hour of a day
Figure 4 Circadian distriburion of the onset of paroxysmal atrial fibrillation-comparison between structural (SHD) and non-structural (NHD) heart diseases.
8:00pm-0:00 am were demonstrated although not significant (p = 0.063). This observation in the NHD group was similar to the outcome obtained by Yamashita et al.27) with a double-curve with peaks at 11:00am to 12:00pm and 10:00pm to 11:00pm. Therefore, it may be an indication that spontaneous cessation of PAF occurs in the morning until around noon and in the late evening.
Dynamic changes in autonomic tone in the
onset and termination of PAF
In recent years, autonomic tone has been easily and noninvasively measured by heart rate variability (HRV) analysis. In particular, the measurement of autonomic tone in humans plays a critical role because reduced vagal and augmented sympathetic activities are reported to be an independent predictor of sudden cardiac death and total mortality in patients with prior myocardial infarction43-45 and chronic heart failure.46,47 On the other hand, HRV analysis is also important for investigating alterations of neural regulatory mechanisms in the genesis of arrhythmias. As a matter of fact, changes in autonomic tone before the onset of spontaneous idiopathic ventricular tachycardia has been investigated and the relationship between autonomic
nerve activity and life-theatening ventricular arrhythmias has been studied in patients with a variety of underlying heart disorders.23)
Coumel31-33) is the first to demonstrate PAF related to vagotonic or sympathetic activity from initiation and its relation to fluctuation in autonomic tone, but his classification was done from clinical observations without HRV analysis. Table 2 shows authors, underlying heart disease, drugs administered and outcome of HRV analysis in reviewed literature of studies on HRV before the onset of PAF. Huang et al.28) classified 3 subtypes according to the HRV analysis before the onset of PAF: onset of PAF accompanied with increased high-frequency component (HF) and decreased low-frequency component (LF)/HF ratio before the PAF onset designated as the vagal type; decreased HF component and increased LF/HF ratio designated as the sympathetic type; and other episodes which did not belong to vagal or sympathetic type designated as the non-related type. As a result, vagal type was predominant (41/63 episodes, 63.5%) in NHD PAF group (idiopathic PAF) whereas sympathetic type was predominant (39/58 episodes, 67.2%) in SHD PAF group.
Other studies also demonstrated strong connection between an increase in HF or increased vagal
Figure 5 Circadian distribution of the termination of atrial fibrillation: comparison between structural (SHD) and non-structural (NHD) heart diseases.
activity with nighttime initiation PAF, and that of an increase in LF/HF and an decrease in HF or increased sympathetic activity with daytime initiation PAF in patients in a drug-free state with structurally normal heart.34,38) Fioranelli et al.35) investgated changes in sympatho-vagal balance only within 5 minutes before the onset of PAF in patients in a drug-free state without structural cardiac diseases. They observed two types of abrupt changes in sympatho-vagal balance even in the last 5 minutes preceding an episode of PAF; Type A with an increase of LF and LF/HF and a decrease of HF, and Type B with a reduction of LF and LF/HF and an increase of HF.
An increase in vagal tone reportedly shortens the atrial refractory period, augmentation of heterogeneity of refractoriness and facilitate intraatrial reentry.48) Strengthened sympathetic tone is likely to induce automaticity and delayed afterdepolariza-tion.49) Therefore, any changes in individual vagal or sympathetic tone, or sympatho-vagal imbalance may initiate PAF.
It is widely recognized that premature beat foci located at or near the orifice of the pulmonary veins are often involved in the genesis of PAF.50-52) Zimmemann and Kalusche36) analyzed dynamic changes in autonomic tone preceding the onset of
PAF with proven pulmonary vein origin. They concluded that episodes of focal PAF originating from pulmonary veins are mainly dependent on variations of autonomic tone, with a significant shift toward vagal predominance before PAF onset.
Bettoni and Zimmermann37) analyzed dynamic change in autonomic tone preceding the onset of PAF by measuring HRV in a total of 147 episodes of the arrhythmia; the 24-hour period, one hour before the PAF onset and the 20 minutes before PAF divided into four 5-minute periods. Their observations were interesting with regards to suggesting a primary increase in adrenergic tone followed by a marked modulation toward vagal predominance before the onset of PAF. The reason for this interpretation was that the LF/HF ratio showed a linear increase until 10 minutes before the arrhythmia, followed by a sharp decrease immediately before it. They also demonstrated a significant alteration in SDANN (standard deviation of RR intervals) and SDNN (standard deviation of the averages of RR intervals for 5 minute segments) parameters of time domain analysis of HRV, and an increase in atrial ectopy in the minute before the PAF onset in more than 50% of patients. Increased frequency of atrial premature beats before the onset
Table 2 Literatures reviewed from investigation of heart rate variability before the oneset of paroxysmal atrial fibrillation (PAF).
Authors
n ... , Nos. of Mean . un
Publish Underlying HD
patients age . Drugs
year , , , (Nos. of patients)
(men/women) y.o.
Outcome of HRV analysis
Huang JL, etal. 1998 57 (34/23) 66 ± 22 ^^^i2? not described
a 1 ' absence (30)
A significant increase in HF in majority of idiopathic PAF (41/63 episodes) and a significant increase in LF/HF ratio (39/58 episodes) in prganic heart disease PAF.
presence
( i | di Digoxin,
Herweg B et al. 1998 29 (18/11) 61 ± 13 ^"blockers,
) ,.„ Ca-blockers absence (13)
Contrary to daytime PAF, an increase in vagal activity preceded predomonantly nocturnal PAF, mostly in younger patients with structurally normal heart.
Fioranelli M etal. 1999 28 (20/8) 60± 11 absence (28) drug-free
Abrupt HRV changes In the last 5 minutes before the onset of PAF. One group revealed an Increase In LF and LF/HF, and a decrease In HF suggesting an enhancement of sympathetic drive. The other demonstrated a reduction of LF and LF/HF, and an increase in HF consistent with an enhancement of vagal drive.
Zimmermann M etal. 2001 13 (10/3)
53 ± 5
absence and
pro drug free
ven pulmonary (11 pts, 85%) vein origin (13)
Pulmonary vein origin AF are mainly dependent on variations of autonomic tone, with a significant shift toward vagal predominance before AF onset.
Bettoni M et al.
2002 77 (63/14) 58 ± 12 absence
amiodarone (9 pts), ^-blockers (18 pts), digoxin (10 pts)
PAF onset greatly depends on variations of autonomic tone, with a primary increase in adrenergic tone followed by an abrupt shift toward vagal predominance.
TomitaTetal. 2003 23 (12/11) 65 ± 2 absence drug-free
PAF was preceded by a gradual increases in HF and LF component without changes of LH/HF ratio in the nighttime initiation PAF. Conversly, daytime initiation PAF group demonstrated a significant increase in the LF/HF ratio before PAF without changes in HF.
of PAF may result in an enlarged standard deviation of RR intervals and a subsequent incorrect interpretation as an increase of vagal predominance. Therefore, HRV should be evaluated without preceding atrial premature beats before the onset of PAF. In this regard, we studied PAF patients with NHD in untreated state and having one or zero premature beats every minute for 5 minutes before the onset of the arrhythmia divided into 4 time periods: morning (6:01am to 12:00pm), afternoon (12:01pm to 6:00 pm), evening (6:01pm to 0:00 am) and night (0:01 am to 6:00 am). Figure 6 shows the changes in HF and LF/HF in the evening onset PAF from just before to 120 minutes before the onset of the arrhythmia. Significanly accelerated HF power and reduced LF/HF ratio just before the PAF onset as compared to 5 to 10 minutes before the onset are observed. Interestingly, note that accelerated HF just before the onset of PAF is preceded by a reduction in HF and reduced LF/HF just before the onset of arrhythmia is preceded by a increase in LF/HF ratio
as compared to those parameters 15 to 30 minutes before the onset. These observations may be an indication that the initiation of PAF needs a reduction in vagal tone or an increase in sympathetic tone, or both about 20 minutes before the onset, and then a sharp increase in vagal tone just before the onset, especially in evening onset PAF.
Tomita et al.38) evaluated HRV parameter changes in a 10-minute segments at 60 minutes, 20 minutes, and at time of onset, not only before the onset of PAF but also after its termination in 23 patients with NHD. They showed a significant decrease in HF and LF components after the termination of PAF in group N (n = 14), in which PAF began at night and was preceded by a gradual increase in HF and LF. This group, however, demonstrated no changes in the LF/HF ratio both before and after the arrhythmia. In contrast, group D (n = 9), in which PAF occurred during the daytime, showed a significant increase in the LF/HF ratio before PAF and a decrease in LF and the LF/HF ratio after PAF,
Minutes before the onset of PAF
Figure 6 Fluctuation in HF component power (top panel) and LF/HF ratio (bottom panel) from 120 to 0 minutes preceding the onset of paroxysmal atrial fibrillation (PAF) in evening initiation PAF. One hundred twenty minutes were divided into twenty-four 5-minute segments from 120 to 0 minutes before the onset of the arrhythmia.
but no changes in HF. Consequently they suggested that the autonomic nervous system played an important role in both the initiation and termination of PAF.
By looking at the variety of studies mentioned above, there is a consensus on the onset mechanism of PAF regarding the fluctuation of autonomic tone in patients with NHD. It frequently initiates during the night and is strongly associated with a sharp increase in vagal tone just before the onset. Some studies emphasize that such a sharp increase of vagal tone is often preceded by a reduction of vagal tone or an increase of sympathetic tone. Very few studies are reported on the autonomic initiation mechanism of PAF in patients with SHD. Some studies demonstrate no change in circadian distribution of PAF onset between NHD and SHD patients. In contrast, our SHD patients showed a triphasic circadian pattern with maximum peaks in the late afternoon, between late evening and night, and in the early morning. Therefore, the autonomic onset mechanism in SHD seems very complicated compared to that in NHD. Further study is needed regarding this.
References
1) Muller JE, Stone PH, Turi ZG, Rutherford JD, Czeisler CA, Parker C, Poole WK, Passamani E, Roberts R, Robertson T, Sobel BE, Willerson JT, Braunwald E, MILIS Study Group: Circadian variation in the frequency of onset of acute myocardial infarction. N Engl J Med 1985; 313:1315-1322
2) Muller JE, Tofler GH, Stone PH: Circadian variation and triggers of onset of acute cardiovascular disease. Circulation 1989; 79: 733-743
3) Hjalmarson A, Gilpin EA, Nicod P, Dittrich H, Henning H, Engler R, Blacky AR, Smith SC, Ricou F, Ross J Jr: Differing circadian patterns of symptoms onset In subgroups of patients with acute myocardial infarction. Circulation 1989; 80: 267-275
4) Muller JE, Ludmer PL, Willich SN, Tofler GH, Aylmer G, Klangos I, Stone PH: Circadian variation in the frequency of sudden cardiac death. Circulation 1987; 75: 131-138
5) Willich SN, Levy D, Rocco MB, Tofler GH, Stone PH, Muller JE: Circadian variation in the incidence of sudden cardiac death in the Framingham Heart Study Population. Am J Cardiol 1987; 60: 801-806
6) Rocco MB, Barry J, Campbell S, Nabel E, Cook EF, Goldman L, Selwyn AP: Circadian variation of transient myocardial ischemia in patients with coronary artery disease. Circulation 1987; 75: 395-400
7) Mulcahy D, Cunningham D, Crean P, Wright C, Keegan J, Quyyumi A, Park A, Fox K: Circadian variation of total ischaemic burden and its alteration with antianginal agents. Lanset 1988; 2: 755-759
8) Tsementzis SA, Gill JS, Hichcock ER, Gill SK, Beebers DG: Diurnal variation of and activity during the onset of stroke. Neurosurgery 1985; 17: 901-904
9) Lown B, Tykocivski M, Garfein A, Brookers P: Sleep and ventricular premature beats. Circulation 1973; 48: 691-701
10) Tanabe T, Yoshikawa H, Tagawa R, Furuya H, Ide M, Goto Y: Evaluation of antiarrhythmic drug efficacy using Holter electrocardiographic technique. Jpn Circulation J 1985; 49: 337-344
11) Furuya H, Tanabe T: Circadian and day-by-day fluctuation of supraventricular premature contractions—inclusive of effects of antiarrhythmic agents. Jpn J Electro-cardiol 1986; 6: 253-259 (in Japanese)
12) Pricheet ELC, Smith MS, McCarthy EA, Kerry LL: The spontaneous occurrence of paroxysmal supraventricular tachycardia. Circulation 1984; 70: 1-6
13) Roark SF, McCarthy EA, Kerry LL, Pricheet ELC: Observations on the occurrence of atrial fibrillation and in paroxysmal supraventricular tachycardia. Am J Car-diol 1986; 57: 571-575
14) Irwin JM, McCarthy EA, Wilkinson WE, Pritchett ELC: Circadian occurrence of symptomatic proxysmal supra-ventricular tachycardia in untreated patients. Circulation 1988; 77: 298-300
15) Michiko Sato, Tanabe T: Circadian distribution of the onset paroxysmal atrial fibrillation and paroxysmal supraventricular tachycardia—inclusive of difference between structural and nonstructural heart disease and between younger and elderly patients. Jpn J Electro-cardiol 1989; 9: 243-252 (in Japanese)
16) Kupari M, Koskinen P, Leinonen H: Double-peaking circadian variation in the occurrence of sustained supra-ventricular tachyarrhythmias. Am Heart J 1990; 120: 1364-1369
17) Rostagno C, Taddei T, Paladini B, Modesti PA, Utari P, Bertini G: The onset of symptomatic atrial fibrillation and paroxysmal supraventricular tachycardia is characterized by different circadian rhythm. Am J Cardiol 1993; 71: 453-455
18) Clair WK, Wilkinson WE, McCarthy EA, Page RL, Pritchett ELC: Spontaneous occurrence of symptomatic paroxysmal atrial fibrillation and paroxysmal supraven-tricular tachycardia in untreated patients. Circulation 1993; 87; 1114-1122
19) Lee SH, Chang PC, Hung HF, Kuan P, Cheng JJ, Hung CR: Curcadian variation of paroxysmal supraventricular tachycardia. Chest 1999; 115: 674-678
20) Task Force of the European Society of Cardiology and North American Society of Pacing and Electrophysiol-ogy, Heart rate variability: standards of measurement, physiological interpretation, and clinical use. Circulation 1996; 93: 1043-1065
21) Lombardi F: Chaos theory, heart rate variability, and arrhythmic mortality. Circulation 2000; 101: 8-13
22) Stein PK, Bosner RE, Kleiger RE, Conger BM: Heart rate variability: a measure of cardiac autonomic tone.
Am Heart J 1994; 127: 1376-1381
23) Fei L, Statters DJ, Hnatkova K, Poloniecki J, Malik M, Camm AJ: Changes of autonomic influence on the heart immediately before the onset of spontaneous idiopathic ventricular tachycardia. Am J Coll Cardiol 1994; 24: 1515-1522
24) Tanabe T, Iwamoto T, Fusegawa Y, Yoshioka K, Shiina Y: Alterations of sympathovagal balance in patients with hypertrophic and dilated cardio-myopathies assessed by spectral analysis of RR interval variability. Eur Heart J 1995; 16: 799-807
25) Kopecky SL, Gersh BJ, McGoon MD, Whisnant JP, Holmes DR Jr. Ilstrup DM, Frye RL: The natural history of lone atrial fibrillation. A population-based study over three decades. N Engl J Med 1987; 317: 669-674
26) Rawles JM, Metcalfe MJ, Jennings K: Time of occurrence, duration, and ventricular rate of paroxysmal atrial fibrillation: the effect of digoxin. Br Heart J 1990; 63: 225-227
27) Yamashita T, Murakawa Y, Sezaki K, Inoue M, Hayami N, Shuzui Y, Omata M: Circadian variation of atrial fibrillation. Circulation 1997; 96: 1537-1541
28) Huang JL, Wen ZC, Lee WL, Chang MS, Chen SA: Changes of autonomic tone before the onset of paroxysmal atrial fibrillation. Int J Cardiol 1998; 66: 275-283
29) Viskin S, Golovner M, Malov N, Fish R, Alroy I, Vila Y, Laniado S, Kaplinsky E, Roth A: Circadian variation of symptomatic paroxysmal atrial fibrillation-Data from almost 10,000 episodes. Eur Heart J 1999; 20: 14291434
30) Vincenti A, Brambilla R, Fumagalli MG, Merola R, Pedretti S: Onset mechanism of paroxysmal atrial fibrillation detected by ambulatory Holter monitoring. Europace 2006; 8: 204-210
31) Coumel P: Neural aspects of paroxysmal atrial fibrillation. In:Falk RH, Podrid PJ, eds. Atrial Fibrillation: Mechanisms and Management. New York: Raven Press. 1992: 109-125
32) Coumel P: Paroxysmal atrial fibrillation: a disorder of autonomic tone? Eur Heart J 1994; 15: 9-16
33) Coumel P: Autonomic influences in atrial tachyarrhyth-mias. J Cardiovasc Electrophysiol 1996; 7: 999-1007
34) Herweg B, Dalal P, Nagy B, Schweitzer P: Power spectral analysis of heart period variability of preceding sinus rhythm before initiation of paroxysmal atrial fibrillation. Am J Cardiol 1998; 82: 869-874
35) Fioranelli M, Piccoli M, Mileto GM, Sgreccia MF, Azzolini P, Risa MP, Francardelli RL, Venturini E, Puglisi A: Analysis of heart rate variability five minutes before the onset of paroxysmal atrial fibrillation. PACE 1999; 22: 743-749
36) Zimmermann M, Kalushe D: Fluctuation in autonomic tone is a major determinant of sustained atrial arrhythmias in ptients with focal ectopy originating from the pulmonary veins. J Cardiovasc Electrophysiol 2001; 12: 285-291
37) Bettoni M, Zimmermann M: Autonomic tone variation before the onset of paroxysmal atrial fibrillation. Circulation 2002; 105: 2753-2759
38) Tomita T, Takei M, Daikawa Y, Hanaoka T, Uchikawa S, Tsutsui H, Aruga M, Miyashita T, Yazaki Y, Imamura
might benefit from ptophylactic treatment with amiodarone: A substudy of EMIAT (The European Myocardial Infarct Amiodarone Trial). J Am Coll Cardiol 2000; 35: 1263-1275
46) Ponikowski P, Anker SD, Chua TP, et al: Depressed heart rate variability as an independent predictor of death in chronic congestive heart failure secondary to ischae-mic or idiopathic cardiomyopathy. Am J Cardiol 1997; 79: 1645-1650
47) Boveda S, Galinier M, Pathak A, et al: Prognostic value of heart rate variability in time domain analysis in congestive heart failure. J Interv Card Electrophysiol 2001; 5: 181-187
48) Ninomiya I: Direct evidence of nonuniform distribution of vagal effects on dog atria. Cir Res 1966; 19: 576-583
49) Bayes de Luna A, Bayes Genis A, Guindo J, Vinolas X, Boveda S, Torner P: Mechanisms favoring and triggering atrial fibrillation. Arch Mal Coeur 1994; 87: 19-25
50) Haissaguerre M, Jais P, Shah DC, Takahashi A, Hocini M, Quiniou G, Garrigues S, Le Metayer P, Clementy J: Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 1998; 339: 659-666
51) Jais P, Haissaguerre M, Shah DC, Chouairi S, Gencel L, Hocini M, Clementy J: A focal source of atrial fibrillation treated by discrete radiofrequency ablation. Circulation 1997; 95: 572-576
52) Hwang C, Karagueuzian HS, Chen PS: Idiopathic paroxysmal atrial fibrillation induced by a focal discharge mechanism in the left superior pulmonary vein: Possible roles of the ligament of Marshall. J Cardiovasc Electrophysiol 1999; 10: 636-648
H, Kinoshita O, Owa M, Kubo K: Role of autonomic tone in the initiation and termination of paroxysmal atrial fibrillation in patients without structural heart disease. J Cardiovasc Electrophysiol 2003; 14: 559-564
39) Tanabe T: Holter and echocardiography assessment in long-term management of atrial fibrillation. In 'Educational Session: Holter or Home Monitoring for management of atrial fibrillation. Proceeding of the 12th Congress of the Society for Holter and Noninvasive Electrocardiology (ISHNE). 2007; 57-58
40) Prinz PN, Halter J, Benedetti C, Raskind M: Circadian variation of plasma catecholamine in young and old men: relation to rapid eye movement and slow wave sleep. J Clin Endocrinol Metab 1979; 49: 300-304
41) Turton MB, Deegan T: Circadian variation of plasma cathecholamine, cortisol and immunoreactive insulin concentrations in supine subjects. Clin Chim Act 1974; 55: 389-397
42) Nahum LH, Hoff HE: Production of auricular fibrillation by application of acetyl-beta-methyl choline to localized region of the auricular surface. Am J Physiol 1940; 129: 428-436
43) Kleiger RE, Miller JP, Bigger JT, et al: Decreased heart rate variability and its association with increased mortality after acute myocardial infarction. Am J Cardiol 1987; 59: 256-262
44) Zuannetti G, Neilson JMM, Latini R, et al: Prognostic significance heart rate variability in post-myocardial infarction patients in fibrinolytic era. Circulation 1996; 94: 432-436
45) Malik M, Camm AJ, Janse MJ, et al: Depressed heart rate variability identifies postinfarction patients who
