Atrial Fibrillation and Its Association With Sudden Cardiac Death

Abstract

Evidence is emerging to indicate that atrial fibrillation (AF) is independently associated with an increased risk of sudden cardiac death (SCD). This association has been consistently observed in specific patient subgroups such as patients with myocardial infarction (MI), heart failure, and hypertension, and importantly, in the general population. Data from studies of implantable cardioverter-defibrillator recipients suggest that the rapid and irregular rhythm of AF and the short-long-short cycles that are highly prevalent in AF increase susceptibility to ventricular tachycardia and ventricular fibrillation. An alternative explanation for the association between AF and SCD includes confounding or mediation by shared risk factors such as coronary artery disease and heart failure. Possible risk factors for SCD in patients with AF include black race, left ventricular hypertrophy, history of MI, and diabetes. Additional research is needed to confirm the inherent proarrhythmic nature of AF, identify patients’ characteristics or clinical conditions that potentiate SCD risk, and define effective SCD prevention strategies for patients with AF. (Circ J 2014; 78: 2588–2593)

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia, and with the aging population its prevalence is increasing over time.^1–3 Additionally, AF is associated with an increased risk of stroke,⁴ heart failure (HF),⁵ and death.^6–11 The Framingham Heart Study reported that AF increases the risk of death by 1.5-fold in men and 1.9-fold in women.⁶ Similarly, the reported Olmsted County, Minnesota experience showed that new-onset AF doubles the risk of mortality.⁷ More recently, the Women’s Health Study revealed that the risk of all-cause death was doubled and cardiovascular death quadrupled by new-onset AF in initially healthy women.⁸

Although the evidence for an independent association of AF with an increased risk of mortality is compelling, there are few data on the causes of death in patients with AF; the most common causes of death in the aforementioned studies were coronary artery disease (CAD) and stroke.^6,7 There is evidence from post-myocardial infarction (MI) or HF patients that AF is associated with an increased risk of sudden cardiac death (SCD).^12–14 However, SCD was not specifically reported in previous population-based studies;^6–8 hence, it is unknown whether AF increases the risk of SCD in the general population. Moreover, because AF and SCD share many common risk factors, such as HF^15,16 and CAD,^17,18 it is possible that any possible association between AF and SCD is confounded or mediated by these shared risk factors.

This narrative review aims to summarize the available information that sheds light on a serious sequela of AF that we are only starting to understand – SCD. We will review the evidence for an independent relationship between AF and SCD in different contexts: recipients of implantable cardioverter-defibrillators (ICDs), patients with MI, HF, or hypertension (HT), and finally, in the general population. We will also examine the determinants of SCD in patients with AF and postulate on the mechanisms underlying this novel and intriguing association.

AF and Ventricular Tachyarrhythmias in Recipients of ICDs

One of the earliest lines of evidence to suggest that AF may inherently increase the incidence of ventricular tachycardia (VT) or ventricular fibrillation (VF) comes from patients who have received an ICD. In a study comprising 229 patients who received ICDs between 1995 and 1999 for secondary prevention of SCD,¹⁹ those who had persistent AF (n=38) at the time of ICD implant experienced appropriate device therapy for recurrent ventricular tachyarrhythmias more frequently than patients who were in sinus rhythm at the time of ICD implantation (n=191) (63% vs. 38%, P=0.01). From multivariable analysis, AF was an independent predictor of both appropriate ICD therapy (relative risk 1.8; 95% confidence interval [CI] 1.2–2.9) and inappropriate device therapy (relative risk 2.3; 95% CI 1.2–4.5). Analysis of device-stored electrograms revealed a higher incidence of short-long-short cycles preceding ventricular arrhythmias in AF compared with patients in sinus rhythm (50% vs. 16%, P=0.002).

Stein et al²⁰ investigated the incidence of “dual tachycardia” [VF or VT preceded by paroxysmal AF or atrial tachycardia (AT)] in patients receiving dual-chamber ICDs for secondary prevention of SCD. Figure shows an example of a dual tachycardia. Overall, there were a total of 2,602 spontaneous VT/VF episodes of which 233 (8.9%) were dual tachycardias. When AT/AF continued between 2 consecutive VT/VF detections, the median inter-detection interval was 15 min. However, when AT/AF terminated (through ventricular therapy or spontaneously) and sinus rhythm persisted for at least 1 min, the median inter-detection interval was 108 h (P<0.001). This observation indicates that the presence of atrial arrhythmias increases vulnerability to VT or VF.

Figure.

Electrogram recording at a sweep speed of 25 mm/s showing an example of dual tachycardia in a patient — AF preceding and concomitant with VT. 1 denotes atrial electrogram, 2 denotes far-field electrogram, 3 denotes ventricular electrogram, and 4 denotes far-field electrogram. The thin red arrows indicate ongoing AF and the thick blue arrow indicates the onset of VT; the ventricular rate accelerates suddenly, accompanied by a change in morphology of the ventricular far-field electrogram. AF, atrial fibrillation; VT, ventricular tachycardia.

The prognostic importance of different types of AF in ICD patients was evaluated in a single-center study at Leiden University Medical Center in the Netherlands.²¹ Data were prospectively collected since 1996 on 913 consecutive patients receiving an ICD for primary or secondary prevention of SCD. At implantation, 73% of patients had no history of AF, 9% had a history of paroxysmal AF, 7% had a history of persistent AF, and 11% had permanent AF. During 833±394 days of follow-up, 117 (13%) patients died, 228 (25%) patients experienced appropriate device discharge, and 139 (15%) patients received inappropriate shocks. Patients with permanent AF exhibited more than double the risk of mortality, ventricular arrhythmias triggering device discharge, and inappropriate device therapy. Patients with paroxysmal or persistent AF did not show a significant increased risk of death or appropriate device therapy, but demonstrated an almost 3-fold risk of inappropriate device therapy.

Collectively, the findings of these 3 studies^19–21 suggest that the presence of persistent or permanent AF may increase susceptibility to VT/VF in patients with (1) history of cardiac arrest from VT/VF, (2) spontaneous or inducible hemodynamically unstable sustained monomorphic VT, or (3) high risk of SCD because of underlying cardiomyopathy.

One caveat, however, should be noted. In ICD-based studies, it may often be difficult to distinguish AF with aberrancy from ventricular tachyarrhythmias.²² To further investigate the relationship of AF to VT/VF in ICD-based studies, exact diagnosis of wide-complex tachycardia is critical.²³ In the following sections, we will outline the evidence that atrial arrhythmias would confer the same increased arrhythmia risk in other patients, such as patients with previous MI or HF.

AF and SCD in Patients With MI

The TRAndolapril Cardiac Evaluation (TRACE) study consisted of 5,983 consecutive acute MI patients admitted to 27 centers in Denmark from May 1990 to July 1992.¹² Of these patients, 1,149 (19%) developed AF during the index hospitalization for MI. During follow-up, there were 1,659 (34%) deaths, of which 536 were classified as SCD. In the multivariable analysis of this study, compared with patients who remained in sinus rhythm, the development of AF during MI was associated with a significant 1.3-fold increased risk of SCD.

Similar observations were made in another study of 505 patients with acute MI enrolled from 3 intensive care units in Italy between 1995 and 1998.¹⁴ Of these patients, 64 (12.7%) developed AF during the first week of hospitalization. After 7 years of follow-up, 217 (43.0%) patients had died. From the multivariable multinomial logistic regression analysis, the presence of AF during the first week of MI was associated with increased odds of SCD (odds ratio, 2.7; 95% CI, 1.2–6.4, P=0.02) but not with HF or other cardiovascular death, and non-cardiovascular death.

AF and SCD in Patients With HF

Middlekauff et al evaluated the relationship of AF to overall survival and SCD in 390 consecutive patients with New York Heart Association Class III or IV advanced HF.¹³ The mean left ventricular ejection fraction was 19±7% and 75 (19%) patients had prevalent AF. After a mean follow-up of 236 days, 98 patients died: 56 (57%) died suddenly, and 36 (37%) died of progressive HF. Overall 1-year survival was 68%, and SCD-free survival was 79%. Overall survival was significantly worse for AF than for sinus rhythm patients (52% vs. 71%, P=0.001). Similarly, SCD-free survival was significantly worse for AF than for sinus rhythm patients (69% vs. 82%, P=0.001). From the multivariable Cox proportional hazards regression analysis, pulmonary capillary wedge pressure on therapy, left ventricular ejection fraction, and AF were found to be independent risk factors for SCD.

In aggregate, the preceding observations suggest that AF is independently associated with increased risk of VT/VF or SCD in specific high-risk patient subgroups: recipients of ICD for primary or secondary prevention of SCD, patients with MI, and patients with advanced HF.^{12–14,19–21} The next 2 sections outline the evidence supporting the association of AF with SCD even in lower-risk individuals: patients with HT and community-dwelling individuals.

AF and SCD in Patients With HT

The relationship of new-onset AF to SCD was evaluated in 8,831 hypertensive patients with electrocardiographic (ECG) left ventricular hypertrophy (LVH) in the Losartan Intervention For Endpoint reduction (LIFE) study.²⁴ The LIFE study was a randomized controlled trial that assigned 9,193 patients with HT and LVH to losartan- or atenolol-based therapy to evaluate the effect of these therapies on cardiovascular outcomes.²⁵ During 4.7±1.1 years mean follow-up, new-onset AF occurred in 701 (7.9%) patients and SCD in 151 (1.7%) patients. From the univariable Cox analyses, new-onset AF was associated with a >4-fold increased risk of SCD (hazard ratio [HR], 4.69; 95% CI, 2.96–7.45; P<0.001). From the multivariable Cox analyses adjusting for statistically significant univariable predictors of SCD, new-onset AF remained associated with a >3-fold increased risk of SCD (HR, 3.13; 95% CI, 1.87–5.24; P<0.001).²⁴ These findings demonstrate that, compared with hypertensive patients with ECG LVH who do not develop AF, those who develop AF have an increased risk of SCD.

AF and SCD in the General Population

To evaluate whether or not incident or new-onset AF is independently associated with SCD in community-dwelling individuals, we recently analyzed data from 2 population-based prospective cohort studies: the Atherosclerosis Risk in Communities (ARIC) study and Cardiovascular Health Study (CHS).²⁶ The ARIC cohort is a bi-racial sample, consisting of 15,792 men and women, 45–64 years of age at baseline (1987–1989), from 4 communities in North Carolina, Mississippi, Minnesota, and Maryland.²⁷ After baseline examination, participants had 4 additional examinations, the last in 2011–2013. In addition, ARIC participants have received annual follow-up calls since the first visit (>90% response rate) collecting information on general health and hospitalizations. The analysis in ARIC was based on data obtained from baseline (1987–1989) through December 31, 2001. The CHS is a cohort study of risk factors for CAD and stroke in older people.²⁸ Between 1989 and 1990, 4 field centers (North Carolina, California, Maryland, Pennsylvania) recruited a total of 5,201 participants aged ≥65 years from Medicare eligibility lists. To enhance minority representation, during 1992–1993, 687 African-American participants were recruited. The analysis in CHS was based on data obtained from baseline (1989–1990 for first cohort and 1992–1993 for second cohort) through December 31, 2006.

For this report on the association of AF and SCD in the general population,²⁶ we analyzed data from 15,439 (baseline 45–64 years, 55% women, 27% black) ARIC and 5,479 (baseline ≥65 years, 58% women, 15% black) CHS participants. We considered only incident or new-onset AF. SCD was physician-adjudicated, defined as CAD death from a sudden, pulseless condition presumed to be caused by a ventricular tachyarrhythmia and non-SCD was defined as CAD death not meeting SCD criteria. In the ARIC study, 894 AF, 269 SCD, and 233 non-SCD events occurred during follow-up (median, 13.1 years). After adjustment for baseline covariates, the respective multivariable HRs (95% CIs) of incident AF for SCD and non-SCD were 3.26 (2.17–4.91) and 2.43 (1.60–3.71) (Table 1). In the CHS, 1,458 AF, 292 SCD, and 581 non-SCD events occurred during follow-up (median, 13.1 years). The respective multivariable HRs (95% CIs) of incident AF for SCD and non-SCD were 2.14 (1.60–2.87) and 3.10 (2.58–3.72) (Table 1). The respective meta-analyzed HRs (95% CIs) of incident AF for SCD and non-SCD were 2.47 (1.95–3.13) and 2.98 (2.52–3.53) (Table 1).

Table 1. Risk of SCD and Non-SCD by AF Status, ARIC and CHS

	Incident AF through 2001, ARIC			Incident AF through 2006, CHS			Meta-analysis
	No	Yes	P value	No	Yes	P value	HR (95% CI)	P value	P value for heterogeneity
SCD
No. of events	238	31		225	67
Person-years	183,086	10,719		58,877	5,582
Crude incidence rate (95% CI)*	1.30 (1.14–1.47)	2.89 (2.00–4.05)		3.82 (3.35–4.35)	12.00 (9.45–15.25)
HR (95% CI)
Model 1**	1 [Referent]	5.40 (3.63–8.04)	<0.001	1 [Referent]	2.51 (1.88–3.33)	<0.001	3.26 (2.58–4.11)	<0.001	0.002
Model 2†	1 [Referent]	3.26 (2.17–4.91)	<0.001	1 [Referent]	2.14 (1.60–2.87)	<0.001	2.47 (1.95–3.13)	<0.001	0.10
Non-SCD
No. of events	189	44		379	202
Person-years	183,086	10,719		58,877	5,582
Crude incidence rate (95% CI)*	1.03 (0.89–1.19)	4.10 (3.02–5.46)		6.44 (5.82–7.11)	36.19 (31.52–41.54)
HR (95% CI)
Model 1**	1 [Referent]	4.66 (3.12–6.95)	<0.001	1 [Referent]	3.55 (2.97–4.25)	<0.001	3.72 (3.15–4.38)	<0.001	0.22
Model 2†	1 [Referent]	2.43 (1.60–3.71)	<0.001	1 [Referent]	3.10 (2.58–3.72)	<0.001	2.98 (2.52–3.53)	<0.001	0.30

AF, atrial fibrillation; ARIC, Atherosclerosis Risk in Communities Study; CHS, Cardiovascular Health Study; CI, confidence interval; HR, hazard ratio; SCD, sudden cardiac death.

*Per 1,000 person-years of follow-up. **Cox proportional hazards model adjusted for age, sex, race, and field center. ^†Model 1 additionally adjusted for baseline covariates: heart rate, smoking status, body mass index, hypertension, diabetes, coronary artery disease, heart failure, ECG-based left ventricular hypertrophy, and use of β-blockers, digoxin, and anti-arrhythmic drugs. In CHS, Model 2 was not adjusted for use of digoxin. (Reproduced from Chen LY et al²⁶ with permission from the American Medical Association.)

To account for confounding by change of covariates over time, we adjusted the main analysis for time-dependent covariates by updating the covariates to the time point just before ascertainment of AF, censoring, or SCD incidence, whichever occurred earlier. We found that AF remained significantly associated with SCD (HR, 2.87; 95% CI, 1.88–4.40; P<0.001) (Table 2).²⁶ To investigate whether shared risk factors (eg, HT, CAD, and HF) could mediate the association between AF and SCD, we conducted another time-dependent regression analysis by updating the covariates to the time point before the end of follow-up. We found that even after adjusting for risk factors that are potentially on the causal pathway, AF was significantly associated with an increased risk of SCD (HR, 2.03; 95% CI, 1.30–3.17; P=0.002) (Table 2).²⁶ In contrast, AF was no longer significantly associated with non-SCD (HR, 1.48; 95% CI, 0.94–2.34; P=0.09) (Table 2). This observation suggests that the association between AF and SCD is only partially explained by the measured shared risk factors, that is, AF per se may increase the susceptibility to VT or VF leading to SCD.

Table 2. Risk of SCD and Non-SCD by AF Using Time-Dependent Covariates, ARIC

	Incident AF through 2001
	No	Yes	P value
SCD
HR (95% CI)
Model 1*	1 [Referent]	2.87 (1.88–4.40)	<0.001
Model 2**	1 [Referent]	2.03 (1.30–3.17)	0.002
Non-SCD
HR (95% CI)
Model 1*	1 [Referent]	2.16 (1.39–3.35)	<0.001
Model 2**	1 [Referent]	1.48 (0.94–2.34)	0.09

Abbreviations as in Table 1.

*Cox proportional hazards model adjusted for age, sex, race, ARIC field center and time-dependent covariates: heart rate, smoking status, body mass index, hypertension, diabetes, coronary artery disease, heart failure, ECG-based left ventricular hypertrophy, and use of β-blockers, digoxin, and anti-arrhythmic drugs. Time-dependent covariates were updated to the time point just before ascertainment of AF, censoring, or SCD incidence, whichever occurred earlier. **Cox proportional hazards model adjusted for the same covariates as in Model 1. However, time-dependent covariates were updated to end of follow-up. (Reproduced from Chen LY et al²⁶ with permission from the American Medical Association.)

That report²⁶ was the first to present evidence for an independent association between incident AF and SCD in the general population. Given the important clinical and public health implications of this association, novel strategies to prevent this devastating outcome of AF are critically needed. To this end, we will review the literature to gain insight into the predictors of SCD in patients with AF and explore the m echanisms underlying this association.

Predictors of SCD in Patients With AF

Information on predictors of SCD in patients with AF can be gleaned from subgroup analyses in the aforementioned studies.^24,26 In the LIFE study, new-onset AF was associated with significantly greater risks of SCD among patients randomized to losartan- vs. atenolol-based therapy and in patients with as opposed to without LVH by Sokolow-Lyon voltage on their baseline ECGs.²⁴ Although the P value for interaction by race (P=0.065) was not statistically significant, new-onset AF in blacks appeared to be associated with a greater risk of SCD (HR, 20.33; 95% CI, 6.02–68.62; P<0.001) than in whites (HR, 4.06; 95% CI, 2.45–6.73; P=0.026). Similarly, in the ARIC study, the risk of SCD associated with AF was higher in black (HR, 5.77; 95% CI, 2.96–11.24; P<0.001) than non-black participants (HR, 2.49; 95% CI, 1.49–4.17; P<0.001) (P for interaction by race=0.02).²⁶

To investigate the determinants of SCD in patients with persistent AF, investigators analyzed data from 522 patients in the RAte Control versus Electrical Cardioversion (RACE) study.²⁹ RACE was a randomized controlled trial that evaluated the cardiovascular outcomes of patients with persistent AF who were randomized to a rate vs. rhythm control strategy.³⁰ After a mean follow-up of 2.3±0.6 years, SCD occurred in 16 patients. Multivariable Cox proportional hazard regression analysis showed that previous MI (HR, 4.9; 95%, 1.8–13.2, P=0.002) and prevalent diabetes (HR, 4.0; 95%, 1.4–11.7, P=0.010) were associated with an increased risk of SCD and the use of β-blockers was protective (HR, 4.0; 95%, 1.4–11.7, P=0.039).²⁹

In summary, possible risk factors for SCD in patients with AF include black race, LVH, history of MI, and diabetes. However, before extending these findings into the clinical domain, additional work is needed to replicate these observations in other independent prospective studies and to understand the underlying mechanisms.

Possible Mechanisms Underlying the Association of AF With SCD

Unlike Wolff-Parkinson-White syndrome, in which the mechanism of AF-induced SCD is quite clear, in the general population the mechanisms underlying this association are less clear. The crux of the matter is whether or not AF is inherently proarrhythmic in the ventricle (ie, AF in and of itself, given the appropriate substrate, would induce VT/VF leading to SCD), or if AF is acting via another factor or condition (eg, HF) leading to increased SCD incidence. This is a critical issue because if the former is the dominant mechanism, efforts to maintain sinus rhythm in high-risk AF patients may prevent SCD. On the other hand, if AF is merely acting via another factor, efforts to maintain sinus rhythm may not be as effective in preventing SCD.

Indeed, evidence to suggest that AF may facilitate the induction of ventricular tachyarrhythmias abounds in the literature. A rapid ventricular rate during an atrial tachyarrhythmia will directly reduce ventricular refractoriness,³¹ promoting ventricular tachyarrhythmias. In addition, the irregular rhythm of AF leads to short-long-short sequences that may be intrinsically proarrhythmic.³² Evidence for AF facilitating induction of ventricular tachyarrhythmias comes from several sources. Somberg et al reported that VT was induced in 25 of 26 dogs by programmed electrical stimulation of the ventricle only during AF, and not in sinus rhythm.³³ As discussed earlier, analysis of device-stored electrograms has revealed a higher incidence of short-long-short cycles preceding ventricular arrhythmias in AF compared with patients in sinus rhythm (50% vs. 16%, P=0.002).¹⁹ Collectively, these observations suggest that atrial tachyarrhythmias may increase susceptibility to ventricular tachyarrhythmias.

Because AF is also associated with an increased risk of non-SCD, it is possible that the association between AF and SCD is mediated by shared risk factors such as CAD or HF.^26,34 The Oregon-Sudden Unexpected Death Study recently reported that in multivariate analysis without considering HF, AF was a significant predictor of SCD. However, in a model that included HF, the AF-SCD association was no longer significant.³⁴ To further assess this issue of confounding or mediation by shared risk factors, we conducted a secondary analysis in the ARIC study by updating covariates to the end of follow-up. Although the association between AF and SCD was attenuated in the analysis, AF remained significantly associated with SCD after adjustment for factors potentially in the causal pathway.²⁶ This observation suggests that the association between AF and SCD is only partially, and not completely, explained by shared risk factors.

Future Directions

Many questions remain regarding the relationship between AF and SCD. First, more evidence (laboratory, clinical, and population-based) is needed to confirm the inherent proarrhythmic nature of AF. Second, even if AF is shown beyond doubt to be inherently proarrhythmic, in what patients and under what conditions, is the proarrhythmic risk the highest? The answer to this question is important because of the obvious implication of targeting SCD prevention strategies to high-risk AF patients. Third, is the SCD risk only associated with persistent or permanent AF, and not with paroxysmal AF? An affirmative answer would provide additional support for clinical efforts to prevent paroxysmal AF from progressing to more advanced forms of AF. Fourth, more data on absolute risk estimates for the association between AF and SCD in different contexts are needed to provide insight into whether or not, and in whom, to pursue SCD prevention strategies. Finally, if SCD prevention in AF patients is justifiable, what might it entail? Should it be limited to the use of β-blockers as shown to be beneficial in the RACE study²⁹ or be extended to rhythm control strategy by catheter ablation?^35,36 Further, should implantation of ICDs in the highest risk patients also be considered? Answers to the aforementioned questions are critically needed to move this evolving and intriguing field forward.

Conclusions

Evidence is emerging to indicate that AF is independently associated with an increased risk of another important public health problem: SCD. The underlying mechanisms are unclear and may include AF being intrinsically proarrhythmic in increasing susceptibility to VT or VF. A comprehensive understanding of the mechanisms and risk factors that increase AF-associated SCD risk is critically needed to reduce the morbidity and mortality of AF.

Acknowledgments

We thank Mr Barry Detloff for his assistance in manuscript preparation. Drs Chen and Alonso acknowledge receiving grant support from the ARIC study (National Heart, Lung and Blood Instistute, N01 HC55019).

Disclosures

Funding: This work was supported by grant funding from the National Institutes of Health.

References

• 1.    Go  AS,  Hylek  EM,  Phillips  KA,  Chang  Y,  Henault  LE,  Selby  JV, et al. Prevalence of diagnosed atrial fibrillation in adults: National implications for rhythm management and stroke prevention: The AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA 2001; 285: 2370–2375.
• 2.    Miyasaka  Y,  Barnes  ME,  Gersh  BJ,  Cha  SS,  Bailey  KR,  Abhayaratna  WP, et al. Secular trends in incidence of atrial fibrillation in Olmsted County, Minnesota, 1980 to 2000, and implications on the projections for future prevalence. Circulation 2006; 114: 119–125.
• 3.    Murakoshi  N,  Aonuma  K. Epidemiology of arrhythmias and sudden cardiac death in Asia. Circ J 2013; 77: 2419–2431.
• 4.    Wolf  PA,  Abbott  RD,  Kannel  WB. Atrial fibrillation as an independent risk factor for stroke: The Framingham Study. Stroke 1991; 22: 983–988.
• 5.    Wang  TJ,  Larson  MG,  Levy  D,  Vasan  RS,  Leip  EP,  Wolf  PA, et al. Temporal relations of atrial fibrillation and congestive heart failure and their joint influence on mortality: The Framingham Heart Study. Circulation 2003; 107: 2920–2925.
• 6.    Benjamin  EJ,  Wolf  PA,  D’Agostino  RB,  Silbershatz  H,  Kannel  WB,  Levy  D. Impact of atrial fibrillation on the risk of death: The Framingham Heart Study. Circulation 1998; 98: 946–952.
• 7.    Miyasaka  Y,  Barnes  ME,  Bailey  KR,  Cha  SS,  Gersh  BJ,  Seward  JB, et al. Mortality trends in patients diagnosed with first atrial fibrillation: A 21-year community-based study. J Am Coll Cardiol 2007; 49: 986–992.
• 8.    Conen  D,  Chae  CU,  Glynn  RJ,  Tedrow  UB,  Everett  BM,  Buring  JE, et al. Risk of death and cardiovascular events in initially healthy women with new-onset atrial fibrillation. JAMA 2011; 305: 2080–2087.
• 9.    Ohsawa  M,  Okayama  A,  Okamura  T,  Itai  K,  Nakamura  M,  Tanno  K, et al. Mortality risk attributable to atrial fibrillation in middle-aged and elderly people in the Japanese general population: Nineteen-year follow-up in NIPPON DATA80. Circ J 2007; 71: 814–819.
• 10.    Suzuki  S,  Yamashita  T,  Ohtsuka  T,  Sagara  K,  Uejima  T,  Oikawa  Y, et al. Prevalence and prognosis of patients with atrial fibrillation in Japan: A prospective cohort of Shinken Database 2004. Circ J 2008; 72: 914–920.
• 11.    Yamashita  T. Recent mortality and morbidity rates of Japanese atrial fibrillation patients: Racial differences and risk stratification. Circ J 2013; 77: 864–868.
• 12.    Pedersen  OD,  Abildstrom  SZ,  Ottesen  MM,  Rask-Madsen  C,  Bagger  H,  Kober  L, et al. Increased risk of sudden and non-sudden cardiovascular death in patients with atrial fibrillation/flutter following acute myocardial infarction. Eur Heart J 2006; 27: 290–295.
• 13.    Middlekauff  HR,  Stevenson  WG,  Stevenson  LW. Prognostic significance of atrial fibrillation in advanced heart failure: A study of 390 patients. Circulation 1991; 84: 40–48.
• 14.    Berton  G,  Cordiano  R,  Cucchini  F,  Cavuto  F,  Pellegrinet  M,  Palatini  P. Atrial fibrillation during acute myocardial infarction: Association with all-cause mortality and sudden death after 7-year of follow-up. Int J Clin Pract 2009; 63: 712–721.
• 15.    Benjamin  EJ,  Levy  D,  Vaziri  SM,  D’Agostino  RB,  Belanger  AJ,  Wolf  PA. Independent risk factors for atrial fibrillation in a population-based cohort: The Framingham Heart Study. JAMA 1994; 271: 840–844.
• 16.    Cupples  LA,  Gagnon  DR,  Kannel  WB. Long- and short-term risk of sudden coronary death. Circulation 1992; 85: I11–I18.
• 17.    Spain  DM,  Bradess  VA,  Mohr  C. Coronary atherosclerosis as a cause of unexpected and unexplained death: An autopsy study from 1949–1959. JAMA 1960; 174: 384–388.
• 18.    Psaty  BM,  Manolio  TA,  Kuller  LH,  Kronmal  RA,  Cushman  M,  Fried  LP, et al. Incidence of and risk factors for atrial fibrillation in older adults. Circulation 1997; 96: 2455–2461.
• 19.    Gronefeld  GC,  Mauss  O,  Li  YG,  Klingenheben  T,  Hohnloser  SH. Association between atrial fibrillation and appropriate implantable cardioverter defibrillator therapy: Results from a prospective study. J Cardiovasc Electrophysiol 2000; 11: 1208–1214.
• 20.    Stein  KM,  Euler  DE,  Mehra  R,  Seidl  K,  Slotwiner  DJ,  Mittal  S, et al. Do atrial tachyarrhythmias beget ventricular tachyarrhythmias in defibrillator recipients? J Am Coll Cardiol 2002; 40: 335–340.
• 21.    Borleffs  CJW,  van Rees  JB,  van Welsenes  GH,  van der Velde  ET,  van Erven  L,  Bax  JJ, et al. Prognostic importance of atrial fibrillation in implantable cardioverter-defibrillator patients. J Am Coll Cardiol 2010; 55: 879–885.
• 22.    Gouaux  JL,  Ashman  R. Auricular fibrillation with aberration simulating ventricular paroxysmal tachycardia. Am Heart J 1947; 34: 336–373.
• 23.    Chishaki  AS,  Li  FJ,  Takeshita  A,  Chishaki  H. Different features of ventricular arrhythmias and the RR-interval dynamics in atrial fibrillation related to the patient’s clinical characteristics: An analysis using RR-interval plotting. J Electrocardiol 2004; 37: 207–217.
• 24.    Okin  PM,  Bang  CN,  Wachtell  K,  Hille  DA,  Kjeldsen  SE,  Dahlof  B, et al. Relationship of sudden cardiac death to new-onset atrial fibrillation in hypertensive patients with left ventricular hypertrophy. Circ Arrhythm Electrophysiol 2013; 6: 243–251.
• 25.    Dahlof  B,  Devereux  R,  deFaire  U,  Fyhrquist  F,  Hedner  T,  Ibsen  H, et al. The Losartan Intervention for Endpoint reduction (LIFE) in hypertension study: Rationale, design, and methods. Am J Hypertens 1997; 10: 705–713.
• 26.    Chen  LY,  Sotoodehnia  N,  Buzkova  P,  Lopez  FL,  Yee  LM,  Heckbert  SR, et al. Atrial fibrillation and the risk of sudden cardiac death the atherosclerosis risk in communities study and cardiovascular health study. JAMA Intern Med 2013; 173: 29–35.
• 27.   The ARIC investigators. The Atherosclerosis Risk in Communities (ARIC) study: Design and objectives. Am J Epidemiol 1989; 129: 687–702.
• 28.    Fried  LP,  Borhani  NO,  Enright  P,  Furberg  CD,  Gardin  JM,  Kronmal  RA, et al. The Cardiovascular Health Study: Design and rationale. Ann Epidemiol 1991; 1: 263–276.
• 29.    Hagens  VE,  Rienstra  M,  Van Veldhuisen  DJ,  Crijns  H,  Van Gelder  IC. Determinants of sudden cardiac death in patients with persistent atrial fibrillation in the RAte Control versus Electrical cardioversion (RACE) study. Am J Cardiol 2006; 98: 929–932.
• 30.    Van Gelder  IC,  Hagens  VE,  Bosker  HA,  Kingma  JH,  Kamp  O,  Kingma  T, et al. A comparison of rate control and rhythm control in patients with recurrent persistent atrial fibrillation. N Engl J Med 2002; 347: 1834–1840.
• 31.    Denes  P,  Wu  D,  Dhingra  R,  Pietras  RJ,  Rosen  KM. The effects of cycle length on cardiac refractory periods in man. Circulation 1974; 49: 32–41.
• 32.    Denker  S,  Lehmann  M,  Mahmud  R,  Gilbert  C,  Akhtar  M. Facilitation of ventricular tachycardia induction with abrupt changes in ventricular cycle length. Am J Cardiol 1984; 53: 508–515.
• 33.    Somberg  JC,  Torres  V,  Keren  G,  Butler  B,  Tepper  D,  Kleinbaum  H, et al. Enhancement of myocardial vulnerability by atrial fibrillation. Am J Ther 2004; 11: 33–43.
• 34.    Reinier  K,  Marijon  E,  Uy-Evanado  A,  Teodorescu  C,  Narayanan  K,  Chugh  H, et al. The association between atrial fibrillation and sudden cardiac death: The relevance of heart failure. JACC Heart Fail 2014; 2: 221–227.
• 35.    Murakawa  Y,  Nogami  A,  Shoda  M,  Inoue  K,  Naito  S,  Kumagai  K, et al. Nationwide survey of catheter ablation for atrial fibrillation: The Japanese Catheter Ablation Registry of Atrial Fibrillation (JCARAF): Report of 1-year follow-up. Circ J 2014; 78: 1091–1096.
• 36.    Latchamsetty  R,  Morady  F. Long-term benefits following catheter ablation of atrial fibrillation. Circ J 2013; 77: 1091–1096.